tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Indexing: need to truncate pathologically long path elements (would cause add_document error)

Browse Source
This commit is contained in:
Jean-Francois Dockes 2011-02-13 10:07:25 +01:00
parent 1c5fe35b7c
commit bf39719ac3
1 changed files with 7 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/rcldb/rcldb.cpp
View File

@ -815,9 +815,8 @@ private:
// Get one term from the doc, remove accents and lowercase, then add posting
bool TextSplitDb::takeword(const std::string &_term, int pos, int, int)
{
#if 0
LOGDEB(("TextSplitDb::takeword: [%s]\n", _term.c_str()));
#endif
LOGDEB2(("TextSplitDb::takeword: [%s]\n", _term.c_str()));
string term;
if (!unacmaybefold(_term, term, "UTF-8", true)) {
LOGINFO(("Db::splitter::takeword: unac failed for [%s]\n",
@ -929,8 +928,12 @@ bool Db::addOrUpdate(const string &udi, const string &parent_udi,
splitter.curpos = 0;
newdocument.add_posting(pathelt_prefix,
splitter.basepos + splitter.curpos++);
for (vector<string>::const_iterator it = vpath.begin();
for (vector<string>::iterator it = vpath.begin();
it != vpath.end(); it++){
if (it->length() > 230) {
// Just truncate it. May still be useful because of wildcards
*it = it->substr(230);
}
newdocument.add_posting(pathelt_prefix + *it,
splitter.basepos + splitter.curpos++);
}